Mark van der Wilk
@mvdw.bsky.social
Associate Professor in Machine Learning at the University of Oxford.
Interested in automatic inductive bias selection using Bayesian tools.
Interested in automatic inductive bias selection using Bayesian tools.
In other work, we investigate metalearning as a way to implement these ideas. The advantage being that a generative model can directly learn the conditional distribution of interest, without a bottleneck of approximate inference!
For more on that, see bsky.app/profile/anis... 3/3
For more on that, see bsky.app/profile/anis... 3/3
Understanding causes is key to science. Finite observational data alone isn't enough. While Bayes offers a framework to deal with this, the calculations are often intractable. We introduce a method to accurately approximate the posterior over causal structures.
#ICLR2025 🧵1/15
#ICLR2025 🧵1/15
July 14, 2025 at 5:19 PM
In other work, we investigate metalearning as a way to implement these ideas. The advantage being that a generative model can directly learn the conditional distribution of interest, without a bottleneck of approximate inference!
For more on that, see bsky.app/profile/anis... 3/3
For more on that, see bsky.app/profile/anis... 3/3
This does lead to the question, what models should we use, and how should we do inference?
We use a VAE with Gaussian Process mappings (GPLVM), but the idea applies equally to Bayesian NNs, if inference can be made to work! 2/3
We use a VAE with Gaussian Process mappings (GPLVM), but the idea applies equally to Bayesian NNs, if inference can be made to work! 2/3
July 14, 2025 at 5:19 PM
This does lead to the question, what models should we use, and how should we do inference?
We use a VAE with Gaussian Process mappings (GPLVM), but the idea applies equally to Bayesian NNs, if inference can be made to work! 2/3
We use a VAE with Gaussian Process mappings (GPLVM), but the idea applies equally to Bayesian NNs, if inference can be made to work! 2/3